[Ecls-list] Cannot print object #<SWANK-BACKEND package> readably

Juan Jose Garcia-Ripoll juanjose.garciaripoll at googlemail.com
Sun Jun 5 13:26:49 UTC 2011


On Mon, May 30, 2011 at 6:08 PM, Ram Krishnan <kriyative at gmail.com> wrote:

> One piece of context here is that the embedded ECL that Terje is using in
> the iPhone Simulator build (or an iPhone device build) doesn't have the
> regular compiler, only the bytecode compiler. I tried a quick test with the
> attached patch on swank-loader.lisp (CVS HEAD), which changes the
> swank-loader to simply load the swank packages instead of compile/loading
> them and it works around the "Cannot print object readably" error.
>

The bytecodes compiler was not using the new toplevel compiler which has
been extended to compile files. The changes w.r.t to form compilation are
not trivial and involve among other things: externalizing constants,
replacing complex constants with MAKE-LOAD-FORM expressions, interpretation
of EVAL-WHEN statements, fixes in DEFMACRO to expand using EVAL-WHEN...

I have been working on this during the last days. The CVS tree now contains
a different version of the bytecodes compiler that is now capable of
producing files and seems to work with Slime. The only thing is that slime
has to be changed to work with the bytecodes compiler. The patches are
attached.

Nevertheless, consider this all to be experimental. This is a new tool and,
as usual, it will need some fine tuning and polishing.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20110605/af673687/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: slime-ecl-bytecmp.diff
Type: application/octet-stream
Size: 1147 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20110605/af673687/attachment.obj>


More information about the ecl-devel mailing list